Version of 16/01/03 ________________________________________________________________________________________________

2

1. INTRODUCTION

On May 2002, the OECD Council at Ministerial Level, stated that:

"The scope for financial crime has widened with the expansion and increased integration of financial markets. Money laundering, terrorism financing and tax crime have all changed in both nature and dimension. Today the potential for financial abuse can threaten the strategic, political and economic interest of sovereign states. Widespread financial abuse undermines the integrity of the international financial system and raises new challenges for policymakers, financial supervisors and enforcement agencies. In certain jurisdictions such abuse may go so far as to undermine the democratic basis of government itself." And then:

"Poorly regulated financial markets not only open up new opportunities for financial crimes but also threaten the stability of the international financial system. As new technologies reduce the importance of physical proximity to major on -shore financial centres, so a new generation of Offshore Financial Centres (OFCs) have emerged. Remote jurisdictions bereft of natural resources and too isolated to benefit significantly from the global economy have established OFCs characterised by strict bank secrecy, criminal penalties for disclosure of client information and a policy or practice of non-co-operation with regulatory, supervisory and law enforcement agencies of other countries. This new generation of OFCs has succeeded in attracting brass plate banks, anonymous financial companies and asset protection trusts".

Do be precise, offshore banking is a generic term that is applied to a variety of financial centres, offering a wide range of services to their customers in a loose regulatory, supervisory, accounting and tax environment. In the most general sense this term should also apply to activities such as trust business, financing vehicles and other corporate vehicles. It is common for these centres to benefit from bank secrecy laws. Offshore banking centres (OFCs) have been used for decades by corporate entities to reduce their tax burden through complex tax planning strategies and by individuals for tax avoidance and evasion. Similarly many of these same centres could be utilised for money laundering.

Therefore, in order to analyse the relationship between the terrorist and organised crime

finance, on one side, and the design of financial regulation, on the other side, we prefer to introduce the definition of Lax Financial Regulation (LFR) countries. An LFR country is a jurisdiction in which the features of the financial regulation increase the probability to offer money laundering services, utilised by the terrorist and criminal organisation1.

The role of LFR countries in international money laundering schemes has long attracted the attention of policy makers. Virtually all initiatives aimed at combating money laundering, both at the domestic and international level, tackle the issue. In the aftermath of “September 11th,” growing attention has been paid to the role of LFR countries in ensuring terrorist financing, adding new perspectives to the debate concerning the initiatives to be taken against such countries.

As a matter of fact, as correctly it has been pointed out2, the first official reaction against terrorism after the September 11 attacks was a financial one: not two weeks has passed since

1 It's important to stress - as Rider (2002) noted - that the laundering process is determined by the need of those agents and organisations seeking to hide or to be disassociated from the wealth, there are important differences between terrorists and conventional criminal organisations. 2 Wasserman (2002)

Version of 16/01/03 ________________________________________________________________________________________________

3

the attack when President Bush signed an executive order freezing the U.S. asset of suspect organisations and individuals3.

Policy makers concentrate their attention on the qualitative 4 and quantitative 5 negative effects of money laundering and on the possibility that LFR centres might facilitate the task of terrorist and criminal organisations. Concerns are raised by regulation adopted in LFR centres, that may greatly contribute to launder money of illicit origins.

Two intertwined postulates commonly feature in the debate concerning the international market for money laundering services: a) money laundering is facilitated by lax financial regulation; b) countries that do not co-operate in the international effort aimed at combating money laundering adopt lax financial regulation. The ensuing observation is that non-co-operative countries contribute to the functioning of the international market for money laundering, and therefore to the world social and economic pollution due to terrorism and organised crime.

The Financial Action Task Force (FATF) for the prevention of money laundering, has endeavoured in an initiative aimed at identifying countries that do not co-operate in the global fight against money laundering. Since 2000, the FATF has monitored 45 countries, that, following our definition, can be identified as potential LFR countries. Using a world-wide data set on the main 130 countries - the United Nation member countries are 189 - we can highlight that this 45 countries represent the 8%, respect to the total GDP, the 15%, respect to the overall population, the 25%, respect to the world foreign bank deposits. And, as OECD pointed out, other LFR countries can exist or emerge.

The FATF produces periodic reports on non-co-operative countries and territories (NCCTs) in an international effort to combat money laundering, commonly – if somewhat incorrectly – described as Black- lists. Since June 2000 five NCCT’s lists have been published (June 2000, June 2001, September 2001, February 2002, July 2002) indicating the jurisdictions that fail to conform to the criteria.6

Therefore, the issues related to the link between black finance and lax financial regulation seem to be qualitatively and quantitatively relevant.

Discussions concerning these issues, however, take often as a given the existence of some countries that offer financial services to terrorism and organised crime, via the adoption of Lax Financial Regulation. In other words, the supply of money laundering services is treated as an exogenous variable.

This paper builds on previous work by the same authors,7 taking a different perspective. We start from the assumption that financial regulation may be a strategic variable for countries that aim at maximising revenues produced by money laundering. A country may find profitable the adoption of a of a financial regulation that attracts capitals of illicit origins or destination We argue that LFR countries are structurally different from other countries. More specifically, we will argue that:

3 On the US new legislation on money laundering see BANOUN, CEPHAS and FRUCHTMAN (2002), and RIDER (2002) 4 Policy makers are mainly concerned with two sources of costs stemming from money laundering. Firstly, the possibility of laundering proceeds of crime affects the incentive of a potential criminal. In a world where money of illicit origins cannot be laundered the possibility of linking the capital to the crime reduces the ex ante incentive of the criminal to commit the crime in the first place. At the margin, more crimes will be committed if money laundering is possible. From this perspective, combating money laundering is equal, in the aggregate, to combating predicate offences. Secondly, capitals that are laundered return to the legal financial sector generating serious negative effects: Competition is distorted; the allocative efficiency of the market is undermined. 5 The size of money laundering flows is unknown by definition. The IMF (1998) estimated the laundered funds as 2-5% of global GDP. 6 See paragraph 7. 7 Masciandaro and Portolano (2001)

Version of 16/01/03 ________________________________________________________________________________________________

4

1. the utility function of countries that favour money laundering is positively correlated to the existence of terrorism or criminal activities abroad;

2. the utility function of such countries is not influenced by the negative effects of such illegal activity, i.e., they do not bear the negative consequences of the terrorist and criminal activity.

Our view is that there may be features of a given country that will naturally support the decision to adopt financial regulation that may in fact facilitate money laundering. In so doing, we take a relational approach, on the assumption that it takes two to tango: we treat regulation that can affect the ease with which money of dirty origins is laundered as a product. Within this framework, we focus on the relationship that is established between a given off-shore country and its customers, i.e. terrorist and/or criminal organisations.

We are less concerned with the main product offered by LFC countries to potential launders (i.e., for example, a strict banking regime) and more concerned with the features of the LFR country that help to support the exchange between that jurisdictions, on one side, and terrorist and criminal organisations, on the other side.

These features may be of various nature. Particular attention will be paid, however, to the economic and institutional environment, generally defined.8 We look for features in the legal system as well as for specific rules that help to sustain the relation that LFR countries and terrorist or criminal organisations establish, thus determining the ultimate success of some LFR centres over others.

Looking at the determinants of success in the competition among LFR countries, it is hoped, will help identify which countries are likely to be involved in money laundering. Grasping the factors that determine the success of some countries in the race to the bottom might also prove useful for policy makers in devising the most appropriate countermeasures.9

In examining the factors that may put a given country in an advantageous position over other countries we take an evolutionist perspective.10 These factors need not necessarily be the result of a “conscious” choice of the country. Rather, they need to prove useful in the competition with other countries. The competitive advantage of a country might also be ascribed to the accidents of history, to geographical factors, or even to sheer chance. For example, the language spoken in the country might obviously play a role in the choice made by criminal or terrorist organisations. An evolutionist approach implies that while we expect a great degree of functiona l convergence, different countries may choose different strategies to the same end.

In discussing the possibility that some countries may act in order to maximise profits stemming from money laundering, we make a simplifying assumption, in that we treat single LFR country as a unitary decision agent. The assumption, albeit naïve, is coherent with the goal of the paper, that is to say, an evaluation of the dynamics of competition among jurisdictions, via the identification of “typical” features of LFR count ries.11

8 For example, as defined in NORTH, (1990) institutions include both formal and explicit rules and less formal rules such as norms. 9 Our attention focuses on countries that try to attract proceeds of crime through the offer of financial services to criminal organisations abroad. We leave aside the broader question of the possible role of off-shore centres in generating and facilitating international financial crises. The latter issue has obviously attracted the attention of policy makers. This interest has also been spurred by the ever increasing integration of financial markets, which has increased the threat to financial stability posed by off-shore centres. See ERRICO and MUSALEM, (1999) FINANCIAL STABILITY FORUM. (2000) 10 As defined in ALCHIAN, (1950) and BECKER. (1962) 11 However, we will sometimes try to shed some light on the black box, in order to look at the possible role of interest groups within the off-shore center. Further research may try to write a thorough “public choice” history of the confrontation that we expect to take place in the political arena within each off-shore country

Version of 16/01/03 ________________________________________________________________________________________________

5

Finally, some remarks on our interdisciplinary methodology. On the economic literature point of view, we clearly follow the classic intuitions of the new political economy, basing our work on the three hyphoteses: the definition of the regulation policy is not exogenous, as in the conventional economics, but endogenous; that policy is not determined maximising a social welfare function, but taking account the political cost - benefit analysis12; the policymaker optimisation is constrained and influenced by the institutional design13 .

We are also indebted toward a strand of literature – usually associated with the “law and economics” movement – that we deem to be strictly, although indirectly, related to the subject matter of our research, i.e. the literature on the competition for corporate charters among the American States that compose the Union. More specifically, we apply in a novel area the approach developed by authors that have tackled the issue in the “transaction cost economics” tradition14.

The importance of institutions is the common element of both our economic and law references. Therefore it is quite natural to acknowledge the suggestions of the recent Law, Endowments and Finance literature15. In fact the importance of the institutional determinants of different national financial structures originates two different body of works: the law and finance theory16, more focused on the legal traditions, and the endowment theory17, more concentrated on the geography/disease endowments. In our work we try to consider both the tradition and natural endowments in determining the degree of laxity.

The paper proceeds as follows. In the first part (paragraphs 2-4) we explore, from the more general point of view, the possible determinants of success of a given LFR country in the market for money laundering. In other words, we try to examine the conditions under which becoming a LFR country can be “convenient” for a given country: what are the geographical, institutional, and economic features that increase the probability that a given country become an LFR country?

12 For the new political economy see DRAZEN (2000) and PERSSON and TABELLINI (2000). 13 See for example, GRILLI, MASCIANDARO and TABELLINI (1991). 14 See ROMANO, (1985), (1993). (1999). To be sure, the situation we examine is not directly comparable with the one examined by American corporate law scholars. The most obvious difference is that competition among the fifty American states takes place under the eye of Federal authorities, namely the Federal Government and the Supreme Court. Especially the latter has shown remarkable attention to the need to reduce the externalities produced by the states. On the other hand the results of a lively debate - dating almost 30 years – allow us to grab fundamental insights even in the context we deal with. The results of such a strand of literature help to develop a theoretical framework of analysis whose application to the subject with which we are concerned appears promising. The circumstance that competition among states takes place in completely different environments – be it the United States, the European Union, or the international market for money laundering services – does not obliterate the idea that competition is likely to respond to the same logic.

Indeed, what started out in the mid 1970s as a purely theoretical debate, evolved over the years into a feast of empirical studies. Measuring the impact of competition on the value of listed companies allows testing the validity of the theoretical conclusions. Despite the sometimes mixed evidence, that literature has gained a solid hold on the dynamics of competition among jurisdictions. More specifically, there appears to be a certain degree of consensus on when and why competition will evolve into a race to the bottom or to the top. We expect empirical research on competition among off-shore countries to be extremely difficult. Obvious factors predict an almost complete lack of information: Parties to money laundering schemes do not publish reports on the success of their operations. By contrast, listed companies supply a goldmine of data for financial economists to measure the impact of the different actions taken by the actors involved. The precision reached by event or accounting data studies does not appear duplicable in the context of the competition among off-shore centres. We therefore have at our disposal an analytical framework whose reliability has been thoroughly tested on the field. 15 See BECK, DEMIRGUC-KUNT and LEVINE (2002). 16 LAPORTA, LOPEZ-DE-SILANES, SHLEIFER and VISHNY (1998). 17 ACEMOGLU, JOHNSON AND ROBINSON (2001).

Version of 16/01/03 ________________________________________________________________________________________________

6

The second part of the paper (paragraph 5) provides a formal theoretical analysis and description of the possible dynamics that govern the policy maker decision to design the financial regulation in order to enter the market for money laundering.

In the third part (paragraph 6) we empirically verify whether the theoretical features of a LFR country are consistent with those of the countries in the FATF’s analysis of effective and potential NCCTs. The empirical evidence – as we shall see – is consistent with the hypothesis. However, we may be ever consider the possibility of facing both Type I and Type II errors. On the one hand, there may be de facto LFR countries that are not included in the FATF analyses; on the other hand, the evidence shows that the NCCTs included in the lists are not entirely homogeneous.

Both observations appear to be important for an overall assessment of the adequacy of the international response, in the conclusive paragraph 7.

2. LAX FINANCIAL REGULATION: KEY CONCEPTS

As already noted, we treat regulation that can affect money laundering as a product, with a demand and supply schedule. But whose demand schedule is driving the system?

Assume that the policy maker in a given country has not yet decided the direction that it will impose on its financial regulation, with specific regard to money laundering. The policy maker may thus decide to implement a regulation that creates serious obstacles to money laundering, and thus to terrorism and organised crime, or it can decide – at the other extreme – to make the opposite choice, devising a lax regulation that facilitates money laundering.

Money laundering generates costs as well as benefits for the parties involved. The costs for society depend on the circumstance that more predicate offences will be committed by terrorist or criminal organisation if money laundering is possible and on the possible negative impact that money laundering will have on the financial system18. The benefits of money laundering accrue, first of all, to terrorist and criminal organisations, that can employ the proceeds of crime avoiding the threat of being prosecuted for predicate offences (money laundering strictu sensu), or that can use legal capital in order to finance illegal activities (money dirtying).

On the other side of the transaction, money laundering offers to the host country the possibility to earn a "commission" in exchange for its services, what we can call the expected national benefits due to lax financial regulation.

Four different categories of actors potentially interested in the regulation can be identified: a) the policy maker; b) terrorist and criminal organisations, deriving utility from the possibility of laundering money; c) those who bear the costs of money laundering; d) the financial community and in general the citizens that received benefits from the inflow of foreign black and grey capitals. Starting with the latter, it does not appear easy to predict which side will the financial community take. In general, we can think that the utility function of financial intermediaries does not appear to be affected by whether profits stem from legal or illegal financial activities (pecunia non olet), if we think that they simply maximise the expected revenues, and that, given the asymmetric information issues, they are not able to distinguish clearly the customers' nature, legal or illegal19 The interests of b-d) and c) are obviously incompatible, as the gains of the former depend on the loss of the latter; a) appears to be caught in the middle, having to decide which demand schedule to follow.

18 For an economic analysis of money laundering as multiplier of the economic and financial impact of criminal organisations see MASCIANDARO (1999) and (1998). 19 For an economic analysis of the role of banking and financial intermediaries in the money laundering process see MASCIANDARO (1996) and JOHNSON and LIM (2002).

Version of 16/01/03 ________________________________________________________________________________________________

7

Note that we are not assuming that b) and c) are necessarily based outside the country where the policy maker we are concerned with is based. This is not an assumption, but rather the consequence of our line of argument. As with all policy issues, as long as the costs and benefits of a decision fall within the boundaries of the area of influence of the policy maker, we expect to have an efficient decision. Policy makers in countries where crime or terrorism is pervasive will tend to bear at least some of the costs associated with a decision to favour money laundering.

Countries where organised crime or terrorism is pervasive could appear to play a minor role in the offer of black or grey financial services at the international level, because they are sensitive to terrorism and crime expected national costs This might be so because the widespread presence of organised crime or terrorism in the country increases for the policy maker the costs of a regulation that favors money laundering.20

Citizens will bear the costs of the decision and will hold the policy maker responsible. Entering the international market for money laundering services has a greater potential for countries that are immune from terrorist or criminal activities. Such countries will almost by definition be able to externalise the costs associated with the increase of predicate offences.21 A negative correlation between crime rate or terrorist episodes in the country and the role played in the offer of money laundering services appears likely.

As a result of this process, some countries which do not bear the costs associated with money laundering become predisposed to adopt a lax regulation that facilitates money laundering. The other side of the coin is that both criminal organisations and those who bear the costs stemming from money laundering will “naturally” tend to be situated in countries other than the one where the regulation is adopted.

We have thus limited our attention to policy makers that are based in countries other than the ones in which the other actors potentially interested in the regulation are based. From this starting point, the confrontation between those who benefit from money laundering and those who suffer from money laundering is almost a “win win” game for criminal and terrorist organisations. Organised crime and terrorist experiences huge asymmetrical organisational advantages over those who bear the costs of money laundering. A small and powerful group faces a large and dispersed group, thus making the outcome predictable.22

To be sure, money laundering regulation could be opposed, and is indeed opposed, by the political authorities that represent the public interest. The dispersion of the costs, however, makes money laundering a low salience issue for the public, and consequently quite low on the 20 We are here leaving aside the possibility of corruption or even mere lobbying by groups interested in having a regulation favorable to money laundering. Through corruption, organized crime might be able to urge the adoption of legislation that facilitates money laundering. We believe this possibility to be less important than it may appear at first glance. For reasons that are developed infra in paragraph 3, a corrupted state will find it difficult to make a credible commitment not to expropriate the assets of illicit origins. 21 These countries will still be exposed to the other source of costs above identified, i.e. the distortion of the functioning of the financial market. This source of costs, however, can be controlled through “ring fencing” practices. Off-shore centers might thus try to build a Chinese wall that insulates its financial system from the effects of involvement in money laundering schemes. For example, a regime favorable to money laundering might explicitly or implicitly exclude residents from taking advantage of its benefits. Conversely, “firms” which benefit from a given regime may be explicitly or implicitly prohibited from operating in the domestic market. Both of these provisions would ensure the off-shore center that criminal organizations that aim at benefiting from the regime do not “reside” in the off-shore center. A similar goal is served by multi-tiered licensing systems. Under such a system, an off-shore center offers two rather different licenses to financial intermediaries, a “restricted” and an “unrestricted” license. A typical multi-tiered regime states that restricted licensees may not engage in transactions with residents inside the off-shore center. They may not collect deposits or even make certain investments. Similar restrictions may also apply to the ability of restricted licensees to solicit funds from the general public. The raison d’être of rules of the type described above is easily perceived. They aim at generating externalities, or more precisely, at avoiding the internalization of costs associated with money laundering. 22 See OLSON, (1965) for a classical exposition of the dynamics of collective action.

Version of 16/01/03 ________________________________________________________________________________________________

8

political agenda. The man on the street simply does not feel the bite of money laundering, and political actors will act consequently.

These hyphoteses - i.e. the role of expected national benefits and of terrorism and crime expected national costs - will be formalised in the analytical part of the paper and tested in the empirical one.

The above described line of argument will help us in the following paragraphs to shed light on the explanations for the casual observation that LFR countries tend to be small, most often they also tend to be islands. In explaining the determinants of the size of “political units”, economic historians have focused on the pressure that are generated by the need to internalise costs associated with economic activities. For example, North and Thomas explained the growth of nation states in the middle age with the need for “political units” that may internalise the costs required to support the expansion of trade.23

In the competition among “political units”, only those that were able to grow were able to exploit the opportunities from the expansion of trade. A converse effect appear to be at work when we come to LFR countries. The need to externalise costs associated with money laundering generates pressures that tend to select countries that are better equipped for the job, countries that may keep organised crime or terrorism – and the costs associated to it – outside of the country.

3. OPPORTUNISM , COMMITMENT AND REPUT ATION: LAX FINANCIAL REGULATION AS CONTRACTUAL DEVISE

The paragraph above shows that some countries will be able to externalise the costs associated with lax financial regulations, thus being in an advantageous position vis-à-vis their competitors in the supply of money laundering services. This advantage, however, does not resolve all problems faced by a potential LFR country.

A closer look at the exchange reveals the problems that need to be solved by both parties. For sake of simplicity, we will juxtapose two parties, Criminal (or Terrorist) and LFR country. Consider a single money laundering operation: It is not a simultaneous game. Parties do not exchange at arm’s length a service for a price or a good for a price. It is rather a game in which one party moves first and the other one moves second. That is to say, Criminal or Terrorist has to move first, deciding whether to put his assets into the hands of LFR country or not. Criminal (Terrorist) is, therefore, a particularly vulnerable consumers of LFR country’s products. Assume that Criminal (Terrorist) decides to move. Once the capitals are in the domain of LFR country, the latter moves. The agreement is to the effect that the LFR country will launder the assets, keep a commission for this services, and then return the laundered assets to Criminal (Terrorist). By so doing, LFR country will earn the commission, a fraction of the overall amount of the assets. LFR country, however, may choose another strategy. It can decide not to co-operate and rather to appropriate the assets. By definition, this strategy implies that the pay off to LFR country will be bigger than if it decided to co-operate.24 In game theoretical jargon, the strategy of non co-operation is a strictly dominant strategy: LFR country will always decide to appropriate.25

There is, in short, the threat of ex post opportunistic switches by LFR countries. In the context of the relation among LFR countries and illegal organisations, appropriation might take

23 “The countries that altered their fundamental institutional arrangements to exploit these opportunities grew, but it was not inevitable that this would occur. For as trade was expanding a need was created for larger political units to define, protect, and enforce property rights over greater areas (thus internalizing some of the costs of long-distance commerce)”, North and Thomas, 1973, at 94. 24 That is to say that the pay-off to Off-shore will be represented by the whole amount of the assets and not by a mere commission. 25 For a classical account of an ex post opportunistic breach by a state see GRANDY. (1989)

Version of 16/01/03 ________________________________________________________________________________________________

9

several forms, running the full gamut from outright taking to merely not responding to the need of keeping financial regulation up to date.

The threat of opportunistic behaviour feeds back into the incentive structure of Criminal (Terrorist). Ex ante the exchange, Criminal (Terrorist) perceives the possibility that LFR country will appropriate the assets; this result unravels thus implying that Criminal (Terrorist) will not put the assets in the hands of LFR country in the first place, in order to avoid a sure loss. The threat of ex post opportunistic behaviour by LFR country translates into the lack of any exchange.

Note that this is a negative result for both parties to the (potential) exchange. Had they been able to co-operate and to realise the exchange they would have both gained, Criminal (Terrorist) from the laundering of the proceeds, and LFR country from the price charged for the service. Despite the potential for a mutually beneficial exchange, the non-simultaneous nature of the game will result in a Pareto move non being made. Unless Criminal (Terrorist) will be assured that LFR country will not behave opportunistically, the exchange will not take place. In other words, the problem is one of transforming a non-co-operative game into a co-operative one.

The problem is exacerbated by the environment in which the relation between Criminal (Terrorist) and LFR country takes place. As it is well known,26 the existence of a state providing an efficient contract law and an efficient enforcement system might manage to help the party to co-operate. LFR country and Criminal (Terrorist), however, bargain in the absence of a superior authority that might perform such function. The threat of opportunistic behaviour is further exacerbated by the circumstance that one of the parties, i.e. LFR country, plays also the role of the enforcer.

But this is not the end of the story. To be sure, Criminal (Terrorist) perceives the threat of ex post opportunistic behaviour by LFR country and will act consequently, but also LFR country perceives that Criminal (Terrorist) will not accept the exchange. Solving the problem is in the interest of both parties: In the absence of a superior authority this becomes the province of endogenous mechanisms of governance.27 More specifically, the parties will have two different but interrelated lines of action, that are capable of transforming a non-co-operative game into a co-operative one.

On the one hand, a non-co-operative game may become a co-operative game if repeated over time.28 We expect LFR country and Criminal (Terrorist) to develop a relational contract, a relation between parties that lasts over a long period of time.29 Rather than playing single money laundering games with different counterparts, parties will have an incentive to play repeatedly with the same opponent. From this point of view, the regulation adopted by LFR country can be regarded as the contractual devise which will govern the relationship as events unfold.30 This is the main reason why in our view, in order to analyse the optimal policy against terrorism finance and organised crime finance we have to concentrate our attention on the regulation architecture.

The second and interrelated line of action aims at reinforcing the relationship. If the main difficulty for LFR country is gaining Criminal’s (Terrorist's) confidence that it will not renege on the agreement, then the issue becomes one of devising a credible commitment not to behave

26 See, for example, COOTER et al. (1999) 27 WILLIAMSON, (1985). (1996) 28 This is only true, however, leaving aside end game problems, i.e. problems that arise when it comes the last period and the parties know that they will not play together anymore. See AXELROD. (1984) 29 The literature on relational contracting is immense; leading contributions are WILLIAMSON, (1979), (1985); (1996) MACNEIL. (1978) 30 Of course, bounded rationality implies that this contract cannot be complete, i.e. it cannot foresee all contingencies. As we shall see in the next paragraph, this limitation has fundamental consequences for the governance of the relationship, mandating the adoption of some form of ex post governance structure.

Version of 16/01/03 ________________________________________________________________________________________________

10

opportunistically. The competition for attracting the black capital will be won by countries that will be able to credibly resolve the commitment problem.

On its face, this observation explains why do not expect to find “corrupted” or “criminal” countries on the supply side of the market for illegal financial services. A Banana Republic, for example, would face immense difficulties in making a credible commitment not to switch course in the middle of the contract. The mere threat that a coup d’état may at any moment overthrow the current regime makes the commitment not credible. Successful states will tend to show a stable political situation.

Extreme cases of political instability aside, however, the way LFR country can commit not to behave opportunistically ex post the exchange depends on its ability to invest in what Williamson terms “transaction specific assets.”31. Transaction specific assets cannot profitably be redeployed outside the original relation. Once a party has invested in such assets, therefore, it has an incentive to continue the relationship, lest it will lose the value of the investment. In other words, LFR country will need to post a hostage,32 i.e. an asset whose value will be lost in case the relationship breaks down due to an opportunistic switch by the LFR country itself.

The most obvious hostage, commonly used in such settings, is reputation. Reputation is often of one and only one use, thus making investments in such asset sunk. While asset specificity is a common feature of reputation, reputation for offering efficient money laundering services, free from the risk of ex post opportunistic behaviour, appears to show an extreme degree of asset specificity. Such reputation requires an intricate set of rules and mechanisms. More specifically, regulation of the financial sector will need to be tuned to reach this goal: The rules concerning banking secrecy, incorporation of business entities, co-operation with supervisory authorities abroad, the duties to report suspicious transactions, to identify and register customers, all these rules will need to be shaped according to the goal of providing an efficient money laundering service.33 Investments need to be made in order to gain a reputation for providing such an efficient service.

Paradoxically, the value of these investments in the closest use is probably negative, rather than being merely zero or little more. A country that has chosen to invest into the financial technology necessary to build a solid reputation as a supplier of first class money laundering services will experience serious difficulties in converting such investments into the next best use. Potential partners of the financial sector situated abroad may be rather sceptical about the credibility of the change in strategy by the LFR country. Should these LFR countries decide to switch course and convert their financial system to lawful uses, they may well start from a negative point rather than merely from zero. Investments will be needed in order to nullify the reputation as a supplier of criminal financial services; once this goal is reached, further investments will be required in order to build from scratch a new reputation, as a supplier of lawful financial services. On this point of view, in the theoretical section we have to investigate the role of such international reputation expected costs.

So far we have examined reputation as a pre-committing device. LFR country however, may use another means to the same results. LFR countries tend to rely on income generated by their financial activity; therefore a change in the policy would be extremely costly for them. This is so because they would experience a severe damage to their level of income.

This observation helps to explain why we can image to find among LFR countries that experience a low level of income and a high dependence from the financial industry revenues: they need to continuously offer financial services to criminal and terrorist organisations because they

31 WILLIAMSON, (1985). (1996) 32 WILLIAMSON. (1983) 33 Of course for some of the subjects mentioned in the text the efficient rule, from the reputational point of view, will be simply no rule, as it is the case with duties to report suspicious transaction reports.

Version of 16/01/03 ________________________________________________________________________________________________

11

want to preserve their level of income or maybe increase it. Furthermore, dependence on income generated by money laundering is yet another committing device.34 A country whose level of income is dependent on the supply of illegal financial services will be committed to offer those services. Such a country might need to fight vigorously in order to preserve its level of income. Countermeasures taken by the international community may urge it to aggressively defend its position. Compare this case with the one of a country whose level of income stems from several sources. The loss generated by the repeal of a policy that attracts capitals of illicit origins will be equal to a fraction of the overall income generated by the state. The incidence of the loss is thus, by definition, less severe.

Dependence on revenues - i.e. the level of expected national benefits - produced by money laundering makes an LFR country a hostage to its own success. In turn, this hostage-like dependence reinforces the bilateral relationship between criminal and terrorist organisations and LFR countries. The former is exposed to the threat of opportunistic behaviour, but the second is exposed to the risk of losing reputation and revenues should it behave opportunistically. Both parties gain from preserving their relationship.

4. LAX FINANCIAL REGULATION AND INSTITUTIONS

Some natural features of certain countries appear to be capable of putting them in an advantageous position in comparison with other countries. The reference to the “natural” character of such features should be intended to imply that they are not only the result of a specific choice of the LFR country. It is rather the other way around. These features are sometimes the result of the accidents of history; in some cases they have even been imposed on the LFR country. Take the adoption of a given legal system, which is virtually always the result of the colonisation of the country by another country that adopted that system. The “natural” features of a winning LFR country will show a sort of “macro” aspect: A low crime rate, the lack of natural resources, the adoption of a common law regime, for example.35

However, once these features have put the specific LFR country down the path of competition with other LFR centres, a demand will arise for institutions that help the LFR country to compete more vigorously. Competitive pressure will urge the adoption of tools that prove useful in the struggle for survival. Starting from the initial positions, a process of refinement through the adoption of newer institutions seems likely. As this process unravels, “micro” institutional devices will be put in place. Interest groups inside the FLR centre will lobby for complementary institutions that increase the value of the existing ones. The institutional environment inside the LFR country will thus be driven, domino- like, by a chain of linked complementary institutions,36 that will add to the survival value of the overall package.

The task of newer institutions appears to be twofold. First, they need to contribute to the overall efficiency of the regulation offered to customers of the off-shore. For example, a strict banking secrecy regime, or rules that protects the anonymity of beneficial owners of accounts.

34 ROMANO, (1993) interprets Delaware’s dependence on income generated by franchise taxes as a pre-committing device. 35 Of course the distinction between “macro” and “micro” institutional devices should not be regarded as one of “quality” but rather of “quantity,” and we use it for mere sake of exposition. With the former we refer to more general and profound institutional features, that tend to characterize a given country with respect to another. By “micro” institutional devices, by contrast, we mean rules that have a more detailed character. 36 The observations in the text are based on GILSON. (2000)

Version of 16/01/03 ________________________________________________________________________________________________

12

Far more interesting for the subject of this paper appears to be, however, the second function of these “micro” institutional devices: Over time, institutional devices that buttress the commitment by the LFR centre are likely to materialise.

How will these pre-committing devices look like? Anything that limits the ability of the LFR country to renege on the agreement with the criminal organisations will do the job. The process of differential survival will select the solutions that serve the pre-committing function. While we expect to observe functional convergence, we also expect to observe a diversity within these devices, whose spectrum is likely to range from a formal and explicit set of rules, for example constitutional rules, to mere norms.

The most obvious example is a supermajority requirement for the repeal of certain pieces of legislation. A rule that states that banking secrecy regulation can be repealed only upon the vote of, say, two thirds of the legislative body makes it more difficult for the LFR country to switch course after Criminal or Terrorism has moved.37 A second device could be a provision to the effect that repeal or modification of a given piece of regulation requires prior approval by organisations representing interests that benefit from regulation sympathetic to money laundering.

For example, the need to obtain the consensus of the bankers’ association or the bar will make it more difficult for LFR country to renege on the agreement. Financial institutions, lawyers, and any other group that makes a business out of the supply of financial services within an international money laundering scheme will fiercely lobby against any initiative that undermines the credibility of the commitment. Even a mere customary norm that requires consultations with the interest groups involved will do, as long as it increases for LFR country the costs of changing course of action and behaving opportunistically.

To be sure, none of these devices is, in itself, a showstopper. Any rule that aims at making the procedure more cumbersome might be repealed thus allowing for the subsequent repeal of the pro-money laundering rule. Take a procedural rule that requires consultations to be held before any modification of rules concerning the financial system may be approved. In anticipation that the financial sector will oppose a change in the regulation that would imply an opportunistic switch, the legislative might first vote to repeal the procedural requirement, and then move on to approve the modification of the regulation. Yet, such a procedure is on its face cumbersome itself. The rule still reaches the goal of increasing the costs of an ex post opportunistic switch, thus helping to fortify the credibility of the commitment.

A last remark: the contractual relation between LFC countries and Criminal or Terrorist is governed, in the first place by the regulation put in place by LFR country. In a world of bounded rationality, however, contracts are hopelessly incomplete. The implicit contract stipulated by LFC country and Criminal or Terrorist is no exception. The regulation cannot specify ex ante all future contingencies. Gaps in contracts are always inescapable, but in the setting we are concerned with the problem appears to be exacerbated by the possibility that one party to the exchange might not reveal all relevant information to the other. The illicit nature of the capitals involved appears to create an incentive for Criminal (Terrorist) to hide some information to LFR country. In fact, there may well be instances in which Terrorist or Criminal will have a clear incentive to disclose false information to LFR country. Beyond the ordinary incompleteness deriving from the costs necessary to write contract clauses,38 there is an increased risk that the contract will suffer from “strategic incompleteness.”39

37 See ROMANO (1993), for the description of a similar provision in Delaware. 38 On which see WILLIAMSON. (1985) 39 Strategic incompleteness is explored in AYRES and GARTNER. (1989)

Version of 16/01/03 ________________________________________________________________________________________________

13

Be it the result of transaction costs or of strategic behaviour, less information translates into more gaps in the contract. The need arises for gap filling devices that allow the party to work out contingencies that were not provided for at the outset.

A country that will be able to offer gap filling devices of superior quality will be in an advantageous position. This shifts the focus of attention towards those features of the legal system that come under severe pressure when it comes to the ex post governance of unspecified contingencies.

We focus on one specific feature that supports the exchange, i.e. the judicial system. The regulation adopted by LFR country fills gaps ex ante, up to the point where the marginal cost and benefit of an added rule are equalised. Remaining gaps will be filled, ex post, by judges. To be sure, the probabilities that a given dispute between illegal organisations and their counterpart inside the LFR centre will go to court might appear low, and indeed it seems reasonable to assume so. At the same time, however, the huge amounts of capitals at stake implies that even with a low probability of a dispute actually going to court, an efficient judiciary might still entail for the parties a high present value. An efficient judiciary works as a last resort mechanism, capable of generating positive externalities on ongoing relations, regardless of whether they actually go to court.

Keeping the quality of regulation constant, therefore, the package that will include the most efficient judicial system will tend to prevail in the competition.

The importance of the judicial power in ensuring the success of an LFR country appears underscored also from a different perspective. The need to fill gaps ex post does not necessarily imply that the gap filling function has to be entrusted to judges. At a purely theoretical level, LFR countries could chose to allocate the gap filling function in the same decisional center responsible for the adoption of the regulation. This solution would probably be infeasible for very practical reasons. When the decision making agent that has written the regulation in the first place is a collective body, say a parliament, entrusting in it the gap filling function would be very impractical. 40 But assume arguendo that the ex ante and ex post gap filling functions are joined. Problems of opportunism aside, this might imply greater familiarity with the issues involved and therefore a higher probability that gaps will be filled in a way consistent with the interest of both parties. By contrast, this advantage is partially lost if the function to decide ex post what the parties involved would have wanted is shifted to a third party.

Putting the threat of opportunistic behaviour back in the picture, however, reveals another advantage of an efficient judiciary. An increased role of judges in filling gaps can also be thought of as one more tool in the “pre-committing” package that a country offers to potential customers. Assigning the task to fill the gaps in the incomplete contract to a judge might also serve another function: À la Madison, 41 fragmenting the powers among many decision making agents helps to ensure that none of them will be able to abuse those powers. An opportunistic switch by the legislative body is likely to require validation by the judiciary. The country that strictly separates the ex ante from the ex post gap filling function will make its commitment more credible.42

40 Every dispute should be examined by a structure whose decision making costs are high, especially if compared with those borne by a single decision making agent, say a judge. The latter can ensure a much higher speed of response, thus being able to handle more issues than the former. Quite obviously, the mere circumstance that in the real world the task to resolve disputes ex post the exchange is indeed entrusted in third parties shows that different solutions would be impracticable. 41 See the famous The Federalist n. 10, MADISON. (1787) 42 This observation obviously paves the way for questions concerning the procedures with which judges are appointed and the possible effects of the procedure on the incentive structure of the judge. For example, life tenure is likely to produce different results from a three year term with the possibility of being appointed again.

Version of 16/01/03 ________________________________________________________________________________________________

14

5. LAX FINANCIAL REGULATION AS ENDOGENOUS VARIABLE: THE MODEL

Now we can try to design analytically the key elements of our approach. In defining the

optimal characteristics of financial regulation aimed at promoting an influx of hidden funds into a given country, we focus on the actions of a national policymaker in what we shall call a Lax Financial Regulation (LFR) country.

Let us assume that this Policymaker is aware that a potential demand for money

laundering exists on the part of one or more criminal or terrorist organisations for a total amount equal to W. We analyse a situation in which the international market of money laundering is demand drive, as it is likely to be in the real world; every potential LFR jurisdiction is a relatively "small country". Each LFR country can define the optimal degree of financial laxity, and then determine the own optimal level of money laundering services supply. The design of financial regulation represent the contractual devise that determine the interrelationships between the country and the illegal organisations 43.

The Policymaker can decide to launder an amount of cash equal to Y, where of course

0<Y<W. In our simple model the decision on the optimal level of money laundering services is equivalent to the choice of the optimal degree of financial laxity.

Calling U the utility function of the Policymaker , it is obvious that the expected utility

from unlaundered profits is zero, whatever their amount:

0)( =− YWU (1) On the other hand, every dollar (or euro?)44 laundered can have a positive expected

value for the Policymaker , since the LFR country can derives benefits from offering financial services that facilitate money laundering. In the preceding paragraphs we showed how a country can derive economic advantages from favouring money laundering. For example, one might hypothesise that the lower the national income and the higher the proportion of that income that depends on the financial industry, the greater will be the propensity to offer money laundering services, all other things being equal. In general let us define those expected national benefits as laxity benefits.

Then the fact that the laundered cash, which we shall indicate with Y, has a positive

expected profitability for the Policymaker may be grasped by imagining that the monetary value B of this benefit is equal to:

mYB = (2)

where m >0 is the expected net rate of return on the money laundering services offered

(i.e. on the degree of laxity) by the LFR country. The inflow of black and grey foreign capital 43 In the model the policymaker choice of the optimal degree of financial laxity is assumed to be equivalent to the decision on the optimal supply level of money laundering financial services. An alternative view should be to consider the degree of regulation laxity one of the possible instrumental variables in order to define the optimal supply of money laundering services; it's a matter of fact however that the link between the money laundering supply and other kind of public policies seem to be logically and empirically more weak. Furthermore, should be easy to model the relationship between laxity and money laundering, considering both random effects and lag effects. 44 For the use of dollar or Euro in the black economy see BOESCHOTEN and FASE (1992), ROGOFF (1997), SINN and WESTERMANN (2001).

Version of 16/01/03 ________________________________________________________________________________________________

15

produces national revenues, increasing the activity of the financial industry and then throughout the traditional macroeconomic multiplier effects45.

If, now, the decision to launder were cost-free, indicating with Y the amount of illegal

funds for which the Policymaker institutes the money-laundering service, it is a simple matter to see that we shall have Y = W. But things are not that simple.

In the first place, an LFR country may be subject to international reputation expected

costs. In the preceding paragraphs we stressed that to be more attractive to criminal or terrorist organisations, a country must make legislative and regulatory choices that increase its credibility as an LFR country. These choices may carry a reputation cost, however, since it cannot be excluded that being an LFR country can cause negative kickbacks, whether in relations with capital, intermediaries and companies sensitive to integrity or with international relations in general. In fact, we have tostress the possibility that under-regulation may be as unattractive for some legal investors as over-regulation46.

Secondly, an LFR country must consider that laundering money means strengthening

inside organised crime or terrorism, i.e. there may be crime & terrorism expected national costs. The Policymaker must first consider the possibility that domestic social damage may derive from the fact that the country is a possible growth engine for criminal organisations. It is obvious, on the other hand, that the less the LFR country registers the actual or potential presence of criminal or terrorist organisations internally, the lower the costs of crime will be perceived.

In our framework we do not separate the crime expected costs from the terrorist

expected costs. From the theoretical point of view, we prefer to stress the Policymaker different sensibility between international expected costs and national expected costs, based on a clear cut different political cost - benefits analysis, that characterised every LFR country. Furthermore, for each LFR country, it should be not difficult to introduce in the expression (3) a specific parameter for each national expected cost factor.

The cost C of offering money laundering for an LFR country will therefore consist of

two parts. First, let us assume that the reputation cost is proportional—according to a parameter c

>0—to the amount of cash it is asked to launder. Secondly, there will be a crime or terrorist cost whose expected value rises as the laundered money increases, for a multiple of the parameter c>0. Let us assume, that is, that for political-electoral reasons the Policymaker of the LFR country, all other things being equal, is more sensitive to the crime and/ore terrorist cost, which can weigh directly on the country's citizens, than to the reputation cost, whose effect on the citizens-voters is probably less perceptible and direct. We therefore have:

YcYC 2γ+= (3)

45 For a macroeconomic analysis of the interrelationships between money laundering, legal and illegal economic sectors see MASCIANDARO (2000). 46 The inflow of legal capital can be assumed as negatively correlated with financial laxity, because of two main effects: a competition effect: in the legal financial sector, competition is distorted and the allocative efficiency of the market is undermined because of extreme financial laxity; a reputation effect: legal customers may fear to suffer a loss of reputation from locating their business in a country highly suspected for money laundering.

Version of 16/01/03 ________________________________________________________________________________________________

16

Lastly, we must consider, as pointed out earlier, that being an LFR country is an

increasing source of economic, political and social risk for the international community. Therefore, when a country decides whether and to what extent to institute a regulatory design that will in essence offer money laundering services, it must consider that this activity is risky, since we assume that the international community might consider it a censurable policy, perhaps even prohibited, and as such subject to sanctions and punitive countermeasures.

Let us assume, therefore, that offering money laundering services can bring with it an

international sanction, whose equivalent monetary value is S, and a probability p that this conduct will be discovered by the international community and thus sanctioned. The probability p can be defined as the degree of technical enforcement of the international sanction. Let us call these risks the international sanction expected costs. In this manner we are able to consider in our model the possibility that the international community define explicit sanction against the LFR country.

The monetary value of the damage from sanctions S against the money- laundering must be at least equal to the value Y of the laundered money. In reality, the damage from a sanction is certainly a multiple, because of the value of the intangible damages related to such a international sanction. So we can assume that the amount of the international sanction is a multiple of the “laundry” volume, equal, for simplicity of computation, to the square of that sum.

And we should also consider that once the crime is discovered, the international

community would apply the sanction with a varying degree of severity, due to a political cost-benefit analysis. The rapidity and procedure for applying the punishment may be variable, affected by national or international structural variables; this severity (or, if you want, the degree of political enforcement)47 with which the sanction is applied can be expressed by variations in the parameter t :

2tYS = (4)

Thus the dilemma of choice facing the Policymaker is the following: if I design lax

regulations that favour the offering of money laundering, and the international community does not sanction it, the benefit for the LFR country is positive, net of the expected cost associated with reputation costs and crime and terrorism risks. If, on the other hand, the LFR country is hit by an explicit international sanction, it will not only sustain the expected costs but will also be damaged by the international sanction. The game is between the Policymaker and the Nature, given that we will assume the "small country" hyphotesis.

47 Rider (2002) noted that the monetary policy international policy is susceptible to political considerations.

Version of 16/01/03 ________________________________________________________________________________________________

17

Graphically:

u(B-C)= u ( )Ycm 2γ−− 1-p p u(-C-T)= u(-cY- 2γ Y-tY2)=-uY(c+ 2γ +tY) W u(0) Having defined the terms of the problem, the Policymaker is thus faced with the

problem of deciding whether and how much to launder, i.e. defining the optimal level of laxity. The optimal policy is not derived by any social utility function, but it is the result of the Policymaker' maximising process, based on his own political cost-benefits analysis. The Policymaker 's expected utility E can now be better specified as:

( )( ) ( )[ ]TCpCBpuUE +−−−= 1)( (5) But since we have define B = m Y and C = c Y + 2γ Y then 5) becomes:

( ) ( ){ } ( )2221 tYYcYupYcYmYpuUE ++−−−−= γγ (6) The linear specification of the function of Policymaker utility tells us that it is a neutral

risk subject. This utility function is consistent with the better economic characteristics in this situation. In fact:

( )[ ] ( ) ( ) 0222 <+−=−−=++−++−=∂∂

mtYuYmtYuYYtYcYYcYYupE

γγ

02 <−=∂∂

upYtE

( ) 01 >−=∂∂

YpumE

International Sanction

No International Sanction

Financial Integrity

Financial Laxity

Version of 16/01/03 ________________________________________________________________________________________________

18

In other words, we find that the utility for the Policymaker , and therefore for the FLR country, declines as the probability of an international sanction and its severity increase, while it increases as the expected return on the money laundering activity increases.

The Policymaker must now determine the optimal level Y* of money to launder,

bearing in mind that the maximum resources available to him, given the potential demand expressed by the criminal or terrorist organisations, amounts to W. Deriving (6) twice for that variable subject to the Policymaker 's decision—to observe the conditions necessary and sufficient for a maximum—we find that:

( )( )[ ] ( )[ ]

( )( )pmcptYu

tpYcpcpcmpuptpYpccmpuYE

−−++−

=−−+−−−=−−−−−−=∂∂

12

2121

2

222

γ

γγγ

uptYE

22

−=∂∂

0<

PROPOSITION ONE: it's possible to define the optimal level of laxity. The function

reaches its maximum at the point

0=∂∂YE

i.e.: ( )( )pmcptY −−++ 12 2γ = 0 which gives us:

( )pt

cpmY

21

*2γ−−−

=

Y* represents the optimal level of money laundering supply services, that is equivalent

to the optimal degree of financial laxity. Let us observe that for Y* > 0 it must be ( ) 21 γ−−− cpm >0 , i.e. the factor of expected benefit from the money- laundering activity,

considering the probability of an international sanction, is greater than the sum of the reputation and crime and terrorism cost factors. Let us define this condition as laxity condition.

It is also possible to define the critical value Y', that marks the limit beyond which it is

definitely optimal for the Policymaker to abstain from offering money- laundering services. Over a certain amount the damage associated with the risk of being punished by the international community is so high that the expected utility is negative, so being an FLR country would not be beneficial. All other conditions being equal, this result depends on the fact that the amount of the sanction is a multiple of the cash to be laundered, so as this value rises the damage from detection

Version of 16/01/03 ________________________________________________________________________________________________

19

of the crime rises more than proportionately. In general this result stresses the importance to have an effective design of the international mechanisms of sanctions.

The critical value Y' must, of course, be compared with the level of potential demand for

resources to launder W. If Y' < W, the amount of resources (W - Y') will be excluded a priori by any laundering decision. If, on the other hand, Y' > W, laundering is potentially advantageous for all the available illegal resources; we must then determine the actual optimal level Y*.

Let us see to what value Y’ corresponds:

( ) ( ) ( ){ } ( )[ ]22211 tYYcYpYcYYrpuUE ++−−−+−= γγ

( ) ( )( )[ ]tpYpcpcmpuYUE −−−−−−= 221 γγ Y=0 therefore ( )UE =0 when

( )( )[ ] ( ) ( )tp

cmptp

cpcpppcmptp

pcpcmpY

222222 111'

γγγγγγ −−−=

−+−+−−−=

−−−−−=

We can evidence the relationships with the structural variables of the model for the

optimal level of laxity. Firstly, the optimal offering of money- laundering will be inversely proportional to the probability of international sanctions:

( )'

21

21

*2

Ypt

cpmY =

−−−=

γ

( )[ ]

( ) ( ) ( )

( )0

2

2222

222222

2122*

2

2

2

2

2

2

2

2

<−+

=

=++−

=+++−−

=−−−−−

=∂

∂

tpmc

pttcttm

pttctptmtmmpt

ptcmptmpt

pY

γ

γγγ

( )0

44*

3

2

24

22

>−−

=−+−

=∂

∂tp

cmtp

mcptpY γγ

Version of 16/01/03 ________________________________________________________________________________________________

20

Therefore, since we have assumed m >c+c2,we find that the first derivative is negative, so the function decreases as the probability of detection increases and the concavity faces upward. i.e. the second derivative is greater than zero. This means that

PROPOSITION TWO : the optimal degree of laxity increases as the degree of technical

enforcement decreases. Y* (p) =0, i.e. it intersects the x-axis at point:

( )0

21

*2

=−−−

=pt

cpmY

γ which means:

( ) 01 2 =−−− γcmp mcm

pcpmm2

2 0γ

γ−−

=⇒=−−−⇒

and we can also say that for

0→p +∞→*Y 1→p t

cY

2*

2γ−−→

As expected, when there are no costs for the LFR country related to its laxity (i.e. c+c2

= 0), that country will abstain from offering money- laundering services (Y*=0) only when the international sanction is absolutely certain (p=1).

As p tends toward zero, the optimal level of laxity for the Policymaker tends to

+∞→*Y , but the Policymaker has available a maximum demand of W, so it must stop with the curve on the probability level at the point where Y* = W .

Let us then find the minimum possible value p can take ( mp ), i.e. at the point where Y*

= W:

( )W

tpcpm

Ym

m =−−−

=2

1*

2γ

tWpcmpm mm 22 =−−− γ 22 γ−−=+⇒ cmmptWp mm

( ) 22 γ−−=+⇒ cmmWtpm

mWtcm

pm +−−

=⇒2

2γ

Version of 16/01/03 ________________________________________________________________________________________________

21

Secondly, the laxity of the LFR country is affected by the severity of the international community in applying the sanction:

PROPOSITION THREE: the optimal degree of laxity increases as the level of

political enforcement decreases.

( )=

−−−=

ptcpm

Y2

1*

2γ

( )[ ]

0412*

22

2

<−−−−

=∂

∂tp

cpmpt

Y γ

Therefore Y* decreases as t increases. When t tends to ∞+ the first derivative is

nullified. What we said about the case where p = p m also applies here. If, in fact, t tends to zero,

we see that Y* tends to ∞+ . But this is not possible, because the maximum level of illegal funds potentially launderable available to the Policymaker is W. Therefore we must also find the minimum value of t (t m) at which Y* = W;

( )W

ptcpm

Y =−−−

=2

1*

2γ

( )

Wpt

cpm

m

=−−−

21 2γ

( ) mWptcpm 21 2 =−−−⇒ γ ( )

Wpcpm

tm 21 2γ−−−

=⇒

The laxity of the LFR country will depend also on the profitability of offering money-

laundering services. PROPOSITION FOUR: the optimal degree of laxity increases as the level of national

benefits increases.

( )pt

cpmY

21

*2γ−−−

= It is a function of the type Y = a x + b where ptp

a2

1−= and

ptc

b2

2γ−−=

Version of 16/01/03 ________________________________________________________________________________________________

22

( )0

21*

>−

=∂∂

ptp

mY

( )0

21

*2

=−−−

=pt

cpmY

γ ( ) 01 2 =−−−⇒ γcpm ( )p

cm−+=⇒

1

2γ

( )

Wpt

cpmY =

−−−=

21

*2

max γ

( )pcWpt

m−

++=

12 2

maxγ

The money- laundering will therefore be non-zero if the profitability lies in the range

] ]max,mmm . Finally, we can then analyse the relationship between the reputation cost of money-

laundering operations and the amount of money to be laundered . PROPOSITION FIVE: the optimal degree of laxity increases as the level of

international reputation costs decreases. As we might expect, the relationship is inverse and equal to:

( )pt

cpmY

21

*2γ−−−

= Y*(c) is a straight line of the type Y = -ax+b

If the reputation cost is extremely high, then Y* = 0. Let us see for what value of c

( )0

21

*2

max =−−−

=ptcpm

Yγ

( )

02

1 2max =

−−−⇒

ptcpm γ

( ) 01 2max =−−−⇒ γcpm

( ) 2

max 1 γ−−=⇒ pmc

02

1*<

−=

∂∂

ptcY

( )W

ptcpm

Y =−−−

=2

1*

2γ

Version of 16/01/03 ________________________________________________________________________________________________

23

( )W

ptcpm m =

−−−2

1 2γ ( ) Wptcpm m 21 2 =−−−⇒ γ

( ) Wptpmcm 21 2 −−−=⇒ γ Lastly, the money- laundering activity of the LFR country will also depend on the

expected crime and terrorism costs, summarised by the parameter c: PROPOSITION SIX : the optimal degree of laxity increases as the level of national

crime and terrorism costs decreases.

( )pt

cpmY

21

*2γ−−−

= ( ) ( )[ ]cpm

ptcpm

Y −−=⇒=−−−

= 102

1* max

2

γγ

0*

<−

=∂

∂pt

Y γγ

01*2

<−

=∂

∂pt

Yγ

if ( )

ptcpm

Y2

1*0

−−=⇒=γ

As the criminal and terrorism risks for its citizens increase, the propensity of the FLR

country to offer money- laundering services decreases. As usual, we can also determine the maximum and minimum values of the parameterc, to which the minimum and maximum of the optimal laundering activity instituted by the Policymaker correspond:

( ) ( ) ( )

( )[ ]Wptcpm

WptcpmWptcpmWpt

cpmY

21

21212

1*

min

222

−−−=⇒

−−−=⇒=−−−⇒=−−−

=

γ

γγγ

Version of 16/01/03 ________________________________________________________________________________________________

24

6. LAX FINANCIAL REGULATION AND NON CO-OPERATIVE COUNTRIES : AN EMPIRICAL ANALYSIS

In the previous sections we theoretically analysed the following hypothesis: a given country may find it advantageous to design its financial regulations, to attract capital of illegal origin, essentially offering money- laundering services. We have designated these states as LFR countries. The LFR countries activities produce benefits for the terrorism and the organised crime.

A country finds it advantageous to become an LFR country because, in defining its objective function, the economic benefits expected from offering money-laundering services are greater than the relative expected costs, associated with internal risk of the development of terrorism and of organised crime, the international risk of loss of reputation, and the possibility of a sanction by the international community. Therefore, the greater the sensitivity of a country to the benefits of money laundering, and the lower its sensitivity to the cost of money laundering, the greater is the probability that it will become an LFR country (Proposition One)

The utility function assumed here must therefore meet these two fundamental requirements: insensitivity to the production of pollution (terrorism and/or organised crime) and a strong sensitivity to the benefit of money laundering services supply.

But what are the economic and institutional characteristics that help define an LFR country? Based on our earlier reflections, being the international contest (i.e. the technical and the political enforcement) constant, we can state that:

* An LFR country will be one that, in terms of economic characteristics, has relatively scant physical resources to spend in international trade, and this is an initial channel of national benefit expected from lax regulation (Proposition Four); * At the same time, an LFR country has the potential for developing financial services, also useful for money- laundering purposes, and this is a second channel of national benefit expected from lax regulation (Proposition Four); * An LFR country also has geographical and social characteristics that shield it to some extent from the risks of terrorism and/or of organised crime and, thus reducing the expected cost of lax regulation (Proposition Six); * An LFR country is relatively indifferent to the expected costs due to international reputation risks (Proposition Five). But, in reality, what category of countries are actually closest to the LFR country model? The answer is immediate, thinking of the activities of the FATF. Formed in 1990, the Financial Action Task Force (FATF) is an intergovernmental body whose objective is to develop and promote policies to combat money laundering, a dangerous process aimed at concealing the illegal income generated by criminal activities. The FATF currently has 29 member countries48 and two international organisations 49. Its membership therefore includes the principal financial centres of Europe, North and South America and Asia.

It is a multidisciplinary body, a fundamental condition for effectively combating money laundering, and possesses the knowledge of experts in legal, financial and economic questions. The need to cover all the aspects of the war against money- laundering is reflected in the Forty

48 Argentina, Australia, Austria, Belgium, Brazil, Canada, Denmark, Finland, France, Germany, Greece, Hong Kong, Ireland, Iceland, Italy, Japan, Luxembourg, Mexico, Norway, New Zealand, Netherlands, Portugal, United Kingdom, Singapore, Spain, United States, Sweden, Switzerland and Turkey. 49 The two international organizations are the European Commission and the Gulf Cooperation Council.

Version of 16/01/03 ________________________________________________________________________________________________

25

Recommendations of the TATF, an instrument which the Task Force decided to adopt and which all countries are asked to follow.

These Recommendations were drafted for the first time in 1990, and revised in 1996 to

incorporate the experience gained in those six years and to reflect the evolution in money laundering. They form the working base for the Task Force and an essential framework of effectiveness in combating money laundering. In particular, since 22 June 2000, the TATF has been publishing a periodic report on non-co-operative countries and territories (NCCTs) in an international effort to combat money laundering: the black- list. The report lays down 25 criteria (Appendix 1) for each country that, if violated, identify the national rules that in each country are detrimental to international co-operation in the fight against money laundering. These criteria are consistent with the Forty Recommendations.

Since June 2000, 45 countries have been monitored (Table 1) , and five black- lists have

been published (22 June 2000, 22 June 2001, 7 September 2001, 1 February 2002, July 2002) indicating the jurisdictions that fail to conform to the criteria: overall, 23 countries have been identified as effective NCCTs at least once (Figure 1).

Analysing the nature of the violations in the NCCTs, country by country, (Table 2) we discover an interesting fact: over 50% of the violations concern deficiencies in financial regulation: no or inadequate regulations and supervision of financial institutions, inadequate rules for the licensing and creation of financial institutions, inadequate customer identification requirements, excessive secrecy provisions regarding financial institutions, and lack of efficient suspicious transactions reporting system.

The principal violations concern the criteria that require that co-operative countries

have an efficient mandatory system for reporting suspicious or unusual transactions to a competent authority, provided that such a system aims to detect and prosecute money laundering (criterion 10), the presence internal regulations of monitoring and criminal or administrative sanctions in respect to the obligation to report suspicious or unusual transactions (criterion 11), and the presence of a centralised financial intelligence unit for the collection, analysis and dissemination of information on suspicious transactions to competent authorities (criterion 25).

The NCCTs therefore tend to violate primarily the criteria related to financial

regulation. It is therefore natural to think that the TATF list of the potential NCCTs is in reality a list of countries that come closest to our theoretical definition of LFR countries. Furthermore, each potential LFR country tends to differ from the others in the number of times it has appeared on the black-list and the number of criteria it has violated. So it may be useful to construct a laxity index, based on this information, to measure the extent to which a given country is lax in its regulations.

Based on the available information, we create a Laxity Index, using a two stage process

(Table 1). First stage: for each country, in every year from 2000 to 2002, each criterion can be satisfied, or can be violated fully or only partially. Therefore for each country a weight can be assigned to each criterion: 0 for compliance, 0.5 for partial violation, and 1 for total violation (Appendix 2). Doing the average of annual criteria violations, we obtain the a preliminary laxity index. Second stage: given the number and the entity of the violations, we must consider that each country can be more or less permanently on the black list. The presence of each countries can potentially range from a maximum of six to a minimum of one. So then, by weighting the respective simple laxity index for each country, we can obtain a final Laxity Index, in which the degree of laxity is found by considering both the number and gravity of the violations of the criteria and the

Version of 16/01/03 ________________________________________________________________________________________________

26

more or less transitory presence of the country on the black list. Therefore for the list of the potential 45 LFR countries we can have the Laxity Index.

Having identified a sample of potential LFR countries, it is possible to perform some

econometric exercises. Using a world-wide data set on the main 130 countries50 (see Table 3 for the list), we do a Probit analysis; the dependent variable is a Binary Probit Variable, that is equal to 1 for the 45 potential LFR countries and 0 otherwise.

The estimated equation is as follows:

t

ttt

EDDCBABinaryLI

εβββββββ

)1()2()1()1()1()1()(

9

876321

++++++=

with Nt K1= where:

A1 = Landuse51; B1 = GDP per Capita52; C1 = Foreign Deposits per Capita53; D1= Democracy Index54; D2 = Commonwealth Country55; E1 = Terrorism and Organised Crime56 Index57.

50 Given the 267 world countries (UN members=180), our 130 countries (BRI sample) represent the 98% of the world GDP and the 90% of the world population. 51 Landuse: This entry contains the percentage shares of total land area for five different types of land use: arable land - land cultivated for crops that are replanted after each harvest like wheat, maize, and rice; permanent crops - land cultivated for crops that are not replanted after each harvest like citrus, coffee, and rubber; permanent pastures - land permanently used for herbaceous forage crops; forests and woodland - land under dense or open stands of trees; other - any land type not specifically mentioned above, such as urban areas. Source: Central Intelligence Agency. 52 Gdp-capita: This entry shows GDP on a purchasing power parity basis divided by population (year 2001). Source: Central Intelligence Agency 53 Fordepositscapita: The data of foreign deposits are derived from reporting as such or calculated by subtracting separately reported data on positions other than deposits from total external assets and liabilities. The only exception is the Netherlands Antilles, which daoes not provide this information separately(year 2001) Source: BRI. The deposit data are then divided by popolatio (year 2001). 54 Democracy Index: www.geocities.com/CapitolHill/lobby/3535/country/list-di.htm 55 Commonwealth Country: dummy variable, equal to 1 if the country is a Commonwealth member, 0 otherwise. 56 Regards the Organised Crime Dummy, it's evident that the drug market dimension is an - indirect and imperfect - indicator of the organised crime problems. It is also true that it's the drug market that has given organised crime its massive resources. It's has been correctly noted that drugs became, during the 70s, a far too profitable and easy trade for even traditional and "conservative" organised crime organisations to ignore. See Rider (2002), pag.17; furthermore, there it's also noted that even terrorist groups, entered into the market, and by so doing became virtually indistinguishable from "ordinary" organised crime. 57 Terrorism and Organised Crime Index: we built this variable summing two separate variables for each country : Organised Crime Dummy = equal to 1 if in the country there are drug production and/or drug markets, 0 otherwise (Source: CIA); Normalised Terrorism Indicator = average number of terrorist episodes in the country (years 1968-91) / max average number of terrorist episodes in a country (yeras 1968-91); the Terrorism indicator therefore ranges from 0 to 1 (Source: Blomberg). Consequently, our Index ranges from 0 to 2 Data Sources; Central Intelligence Agency – www.cia.gov/cia/publications/factbook; Democracy Index – www.geocities.com/CapitolHill/Lobby/3535/country/list -di.htm; Foreign Bank Deposits: Bank for International Settelements – www.bri.org/publ/qtrpdf/r_qa0206.pdf#page=44; ; Terrorism Indicators, see Blomberg B.S., Hess D.G., Weerapana A., Terrorism From Within: An Economic Model of Terrorism, May 2002 and ITERATE Data Set.

Version of 16/01/03 ________________________________________________________________________________________________

27

------------------------------------------------------------------------------------------------------------------ Probit estimates Number of obs = 130 LR chi2(6) = . Log likelihood = -65.094742 Prob > chi2 = . ------------------------------------------------------------------------------ BinaryProbit | Coef. Std. Err. z P>|z| [95% Conf. Interval] -------------+---------------------------------------------------------------- landuse | .0055706 .0037493 1.49 0.137 -.0017779 .0129191 gdpcapita | -.0000673 .0000236 -2.85 0.004 -.0001136 -.0000211 fordeposit~a | 3.40e-06 1.43e-06 2.38 0.017 5.98e-07 6.19e-06 DandTvaria~e | -.4496926 .2352834 -1.91 0.056 -.9108397 .0114545 democracy | -.0010027 .0056341 -0.18 0.859 -.0120454 .0100399 commwealth | .2645573 .2975846 0.89 0.374 -.3186979 .8478124 ------------------------------------------------------------------------------ Marginal effects after probit y = Pr(BinaryProbit) (predict) = .7814722 ------------------------------------------------------------------------------ variable | dy/dx Std. Err. z P>|z| [ 95% C.I. ] X ---------+-------------------------------------------------------------------- landuse | .0016431 .00137 1.20 0.230 -.001043 .004329 63.4623 gdpcapita | -.0000199 .00001 -2.06 0.039 -.000039 -1.0e-06 11403.8 fordepts | 1.00e-06 .00000 9.35 0.000 7.9e-07 1.2e-06 423892 T&Crime | -.1326382 .10291 -1.29 0.197 -.334336 .069059 .627538 democracy | -.0002958 .00166 -0.18 0.859 -.003549 .002958 38.9462 commwealt*| .0743387 .08 0.93 0.353 -.082459 .231136 .276923 ------------------------------------------------------------------------------ (*) dy/dx is for discrete change of dummy variable from 0 to 1

The regression shows that the probability to be a LFR country will depend on economic variables (Proposition One).

Firstly, we have stated that an LFR country will be one that, because of its geographical

and economic characteristics, has relatively scant physical resources to spend in international trade, and this is the first channel of national benefit expected from lax financial regulation (Proposition Four). We note that the probability that a country became a LFR country tend to be higher:

* the more she experiences economic growth problems, measuring those problems in terms of per-capita GDP (significance at 0%) and the level of land exploitation (significance at 13%);

Secondly, we have affirmed that an LFR country has the possibility of developing its

offering of financial services, also useful for purposes of money laundering, and this is a second channel of national benefit expected from lax financial regulation (Proposition Four) . In this regard, we note that the probability that a country became a LFR country tend to be higher:

* the more they have developed the flow of foreign deposits (significance at 1%)

Version of 16/01/03 ________________________________________________________________________________________________

28

Thirdly, we have stated that an LFR country will be one that have no terrorism and/or

organised crime problems (Proposition Six). In the regression we use a joint Index of the terrorism risks and of the organised crime risks. In our approach every national policymaker care about both risks, and a lax financial regulation can benefits in principle either terrorism or organised crime. We note that the probability that a country became a LFR country tend to be higher: * as the degree of terrorism and organised crime risks decrease (significance at 5%).

Finally, the probability that a country became a LFR country tend to be higher: as the

degree of democracy decreases, measured by the Democracy Index (significance at 85%); and if the country is a Commonwealth member (significance at 37%). We have to highlight that we do not find data to test the role of the international reputation sensitivity (Proposition Five)58.

Using the same data set, we can also reach some conclusions on the relationships

between potential LFR countries and offshore centres.

While there is a theoretical presumption that international tax evasion and money laundering through offshore centres should overlap59, there is no need for this to be the case. Map 1 shows the list of countries that are currently not complying with the Guidelines of the Financial Action Task Force. Table 3 show the countries that have been identified by the OECD as engaging in Harmful Tax Competition. As can be seen by comparing the two black lists, not all tax favoured centres are also not compliant with FATF. However, owing to a lax regulatory and supervisory environment banks in those countries are considered as potential targets for money laundering. Similarly a number of “black listed” locations from a tax standpoint are not NCCTs (non-co-operative countries and territories) under FATF.

While this finding may appear to contradict the hypothesis that money laundering and tax

evasion are intimately related, this is not altogether surprising in light of the criteria utilised for constructing the “black lists” for NCCTs and centres engaging in “harmful tax competition”. The members of FATF did not seek to list a series of transactions that could be identified as “money laundering” and then seek to identify centres in which such transactions were more likely to occur. Rather forty criteria basically relating to the degree of “transparency” of the financial system and to the degree of law enforcement were utilised for defining NCCTs.

In drawing up the “black list” FATF was also careful not to avoid a number of complex

issues, such as the definition of “criminal” activity which can vary from centre to centre. This is especially important for “tax evasion” which is treated very differently under the law in various jurisdictions. As a consequence, the emphasis on “non transparency”, degree of supervision and law enforcement has led to a very disparate series of jurisdictions appearing as NCCTs.

In the case of the OECD “harmful tax competition” studies (1998) and EU Report of the

Primarollo Group, the focus was on identifying tax provisions or preferential regimes in some instances of a very specific nature or relating to a specific type of industry (for example, shipping). The focus was on transactions that could lead to the shifting of the tax base and not necessarily to “tax evasion” which entails the concealment of income. The coverage included regimes in OECD countries and chapters were dedicated to a number of specific topics: artificial definition of the tax 58 Obviously we cannot test with a cross-country analysis the role of international economic and political enforcement, that, in the traditional economic policy point of view, are variables no country specific, while , in a new political economy point of view, should be more testable prima facie using country - case studies. 59 See ALWORTH and MASCIANDARO (2002), YANIV (1994) and (1999).

Version of 16/01/03 ________________________________________________________________________________________________

29

base; failure to adhere to international transfer pricing principles; foreign source income exempt from residence country tax; negotiable tax rates or tax base; existence of secrecy provisions; access to a wide network of tax treaties; regimes which are promoted as tax minimisation vehicle; and regimes encourages purely tax-driven operations or arrangements.

There is also another important distinction that can be made between countries engaged in

Harmful Tax Competition and NCCTs. Given the complexity of many money laundering schemes, highly sophisticated intelligence and policing operations may be needed to control and eradicate money laundering. This may be difficult for many small centres, particularly because money laundering by its very nature involves circumventing the rule of law and small financial centres may not have the resources to police these activities. Consequently, the efforts of FATF aimed at improving the workings of the financial systems of small financial centres may at times actually assist the development of “more efficient” tax havens.

Therefore, using the same world-wide data set, we do another Probit analysis; the

dependent variable is now a Offshore Binary Probit Variable, that is equal to 1 for the OECD offshore countries and 0 otherwise.

The estimated equation is as follows:

t

ttt

EDDCBAOffshoreLI

εβββββββ

)1()2()1()1()1()1()(

9

876321

++++++=

with Nt K1= where:

A1 = Landuse; B1 = GDP per Capita; C1 = Foreign Deposits per Capita; D1= Democracy Index; D2 = Commonwealth Country (Dummy variable); E1 = Terrorism and Organised Crime Index; ----------------------------------------------------------------------------- Probit estimates Number of obs = 130 LR chi2(5) = . Log likelihood = -59.028787 Prob > chi2 = . ------------------------------------------------------------------------------ dummyOffShore | Coef. Std. Err. z P>|z| [95% Conf. Interval] -------------+---------------------------------------------------------------- landuse | .0018127 .0038405 0.47 0.637 -.0057145 .0093399 gdpcapita | -3.79e-07 .0000151 -0.03 0.980 -.00003 .0000293 fordeposits | 6.37e-07 5.39e-07 1.18 0.238 -4.20e-07 1.69e-06 T&Crime | -1.325194 .2739108 -4.84 0.000 -1.86205 -.7883389 commwealth | -.3228179 .2960498 -1.09 0.276 -.9030647 .257429 ------------------------------------------------------------------------------ Marginal effects after probit y = Pr(dummyOffShore) (predict) = .31079773 ------------------------------------------------------------------------------ variable | dy/dx Std. Err. z P>|z| [ 95% C.I. ] X ---------+-------------------------------------------------------------------- landuse | .0006402 .00135 0.47 0.636 -.002014 .003295 63.4623

Version of 16/01/03 ________________________________________________________________________________________________

30

gdpcap~a | -1.34e-07 .00001 -0.03 0.980 -.000011 .00001 11403.8 fordep~a | 2.25e-07 .00000 1.06 0.289 -1.9e-07 6.4e-07 423892 d0tdummy*| -.4640011 .07904 -5.87 0.000 -.618913 -.30909 .592308 commwe~h*| -.1094256 .09404 -1.16 0.245 -.29375 .074899 .276923 ------------------------------------------------------------------------------ (*) dy/dx is for discrete change of dummy variable from 0 to 1

It's quite evident that the regression have not explanatory power. The only exception is represented by the Terrorism and Organised Crime Index: We can say that the probability that a country became an Offshore centre tend to be higher :

* as the degree of terrorism and organised crime risks decrease (significance at 0%). In general, therefore, we can reject the hyphotesis that the causes of LFR decisions and

of Offshore activities are the same.

7. CONCLUSIONS

The lax financial regulation problem, favouring the money laundering phenomena, can increase world-wide the risks of terrorism and of organised crime. In this paper we have explored theoretically and empirically the issue.

Theoretically, the degree of financial laxity can be an endogenous variable, determined by the policy maker cost -benefits analysis, depending then on economic and institutional country variables, as the growth level, the role of the financial industry, the reputation sensibility, the absence of terrorism and/or of organised crime, the institutional attractiveness.

Empirically, the empirical analysis does not repudiate the theoretical assumption that countries, that because of scant resources, foreign dependence in the offering of financial services, and absence of terrorism and/or organised crime risks, and perhaps insensitivity to the international community judgement, can derive net expected national benefits from offering laundering services for illicit foreign capital, and therefore can be or become LFR countries.

The empirical relationships developed are probably interesting but not definitive or conclusive. This prompts at least three reflections. In terms of results, we must stress that the potential LFR country display uniform economic and institutional elements, bolstering the significance of the FATF action, but also marked dissimilarities among them. This suggests two indications for designing international policies of prevention and combat. On the one hand, that by modifying their formal rules they do not automatically cease to be LFR countries, since the incentives for laxity in combating the laundering of illicit capital may be very deep-rooted. On the other hand, that the international community can have an impact on those roots through stick-and-carrot policies tailored to each country, precisely because the degree of laxity and its motivations may not be identical in each case.

On a battlefield where reputation is one of the main weapons, policy makers engaged in the fight against international money laundering schemes should be very cautious in taking initiatives that may affect the reputation of the actors involved. A pure “name and shame” approach may even prove counterproductive. Tampering with reputational mechanisms might, at the same time, not only miss the target but also reach the wrong target. First, there is a high risk of false negatives, i.e. of including in a hypothetical list of countries that supply money laundering services countries that are merely engaged in the offer of financial services of superior quality. The costs of

Version of 16/01/03 ________________________________________________________________________________________________

31

such an error appear great. To put it with the Financial Stability Forum, “not all [Off-shore centres] are the same. Some are well supervised and prepared to share information with other centres, and co-operate with international initiatives to improve supervisory practices. But the Survey carried out by the [Financial Stability Forum] indicated that there are serious concerns by onshore supervisors about the quality of supervision in, and degree of co-operation provided by, some [Off-shore centres].”60

Reputation is the basic tool of the trade also for countries that are not involved in money laundering schemes but are merely aiming at attracting capitals from abroad thorough the offer of superior quality financial services. From this perspective, a mistake by the international community that includes the wrong country in the list might cause serious distortions in the competition among jurisdictions. These countries, like victims of friendly fire, will find their reputation in the financial community seriously hampered, to the detriment of their role in the market. In the long run, such types of mistake appear also capable of curbing innovation in the financial sector. Regulatory arbitrage is a powerful force in driving innovation, and the international community should recognise that tinkering with the reputation of the actors involved is a dangerous game.

But even assuming that the international community is capable of effectively singling out LFR country that are indeed involved in money laundering schemes, a cautious approach is still deemed necessary. When the international community points the finger at a given country as a leading supplier of money laundering financial services, it may also be certifying, to the benefit of the country itself, that that country is indeed specialised in that business.

The signalling effect embedded in the “name and shame approach” should not be underestimated. The main difficulty for a LFR country is solving credibly the commitment problem: Then, what’s best for the LFR country than having the international community, not exactly its closest friends, solving that problem with a public statement? Listing should also be regarded as a sort of third party bonding, which is likely to generate two intertwined effects. First, it is capable of cementing the commitment by the LFR country. Secondly, naming increases the transaction specific character of investments in reputation. The inclusion in a list increases the value of the (sunk) investments in reputation. A state that is engaged in money laundering and that finds itself blacklisted will find it even more difficult to switch course and decide to exit the market, thus being encouraged to compete aggressively in the market. The final result does not change much. They still need to move forward.

This is not to say that the international community should not endeavour in listing countries that are involved in the market for money laundering services. Quite to the contrary; what this paper argues, is that a per se “name and shame” approach, separated from other initiatives, equals to a third party seal on the reputation of LFR countries. Names should be named, but only if blacklisting goes hand in hand with other measures that are capable of outweighing the positive effects experienced by the off-shore centre as a result of the inclusion in the list.

Appropriate countermeasures should be grounded on the premise that even the most efficient LFR country will still need, in a globalised world, to be integrated in world financial markets. This implies that no matter how many layers of transactions cover the predicate offence, terrorism or criminal organisations will still need to place that money within the lawful financial sector. This step is necessary, at a minimum, in order to exploit in lawful uses the capitals, once they have been laundered. Money laundering is by definition instrumental to a later use.

With this regard, it should be noted that there is a fundamental feature of the initiative taken by the FATF that appears to be pivotal for its success. The FATF has not limited its initiative to a mere recognition of “non co-operative countries and territories.” FATF member states have

60 FINANCIAL STABILITY FORUM, (2000) Executive Summary, at 2.

Version of 16/01/03 ________________________________________________________________________________________________

32

also applied “Recommendation 21”61 to the countries included in the list. “Recommendation 21” requires a higher scrutiny by financial intermediaries in evaluating the possible suspect nature of transactions with counterparts, including legal persons, based in a country listed as non-co-operative. As a result of the FATF initiative, many countries included in the list have already taken initiatives aimed at overcoming the serious deficiencies observed by the FATF.62

These initiatives need to be evaluated in the medium to the long run, because, for example, some of the enacted laws will need secondary regulations to be put in place to become effective, or, more generally, the initiatives taken at the legislative level will need to be followed by concrete actions. However, it can be argued that the threat of being crowded out by the international community has played a great role in spurring the adoption of the above mentioned initiatives.

The second conclusion that can be reached on the basis of the empirical evidence we have examined, is that we must not exclude the possibility that there are LFR countries not presently included in the FATF monitoring action, perhaps because they are highly effective in bringing their formal rules in line with international precepts, while in their deeds they remain lax in the fight against money laundering. This implies a constant effort on the part of international organisations, particularly the FATF, in updating the criteria and monitoring the countries.

61 See Fatf, (1990). (2000) 62 See Fatf press communiqué of October 5th, 2000.

Version of 16/01/03 ________________________________________________________________________________________________

33

9. REFERENCES

ACEMOGLU, D., JOHNSON S. and ROBINSON J.A. (2001), The colonial origins of comparative development: an empirical investigation, American Economic Review, n.91, 1369-1401.

ALCHIAN, A. A., (1950) Uncertainty, Evolution, and Economic Theory, in 58 Journal of Political Economy, 211.

AXELROD, R., (1984) The Evolution of Cooperation, New York.

ALWORT, J and MASCIANDARO, D., (2002), Tax Evasion, Money Laundering and Offshore Centres: where is the connection?, World Bank International Tax Conference, Brighton, October, 30-31, (forth).

AYRES, I. and GERTNER, R., (1989) Filling Gaps in Incomplete Contracts: An Economic Theory of Default Rules, 99 Yale Law Journal, 87.

BANOUN R., CEPHAS D. and FRUCHTMAN L.D., (2002), US Patriot Act and Other Recent Money Laundering Developments Have Broad Impact on Financial Institutions, Journal of Taxation of Financial Institutions, v.15, n.4.

BASEL COMMITTEE ON BANKING SUPERVISION, (1988) Prevention of Criminal Use of the Banking System for the Purpose of Money-Laundering, Basel.

BECK T., DEMIRGUC-KUNT A. and LEVINE R. (2002), Law, Endowments and Finance, NBER Working Paper, n. W9089.

BECKER, G., (1962) Irrational Behaviour and Economic Theory, in 70 Journal of Political Economy, 1.

BOESCHOTEN W. and FASE M., (1992), The Demand for Large Bank Notes, Journal of Money, Credit and Banking, vol. 24, No 3, 1992. DRAZEN A., (2000), Political Economy in Macroeconomics, Princeton, Princeton University .

ERRICO, L. and MUSALEM, A., (1999) Offshore Banking: An Analysis of Micro- and Macro-Prudential Issues, Working Paper of the International Monetary Fund, n. 5.

FINANCIAL ACTION TASK FORCE, (2000) Review of Non-Co-operative Countries or Territories: Increasing the Worldwide Effectiveness of Anti-Money Laundering Measures.

___________________________, (1990) The Forty Recommendations.

FINANCIAL STABILITY FORUM, (2000) Report of the Working Group on Off-Shore Financial Centres.

GILSON, R. J., (2000) Globalizing Corporate Governance: Convergence of Form or Function, Columbia Law School Working Paper n. 174.

GRANDY, C., (1989) Can Government Be Trusted to keep Its Part of a Social Contract? New Jersey and the Railroads, 1825 – 1888, in 5 Journal of Law, Economics, and Organisation, 249.

GRILLI V., MASCIANDARO D. and TABELLINI G., (1991), Institutions and Policies. Political and Monetary Institutions and Public Financial Policies in the Industrial Countries, Economic Policy, n. 13, 341-376.

HAYEK, F. A., (1960) The Constitution of Liberty, Chicago.

Version of 16/01/03 ________________________________________________________________________________________________

34

INTERNATIONAL MONETARY FUND (1998), Money Laundering. The Importance of International Countermeasures, address by Michel Camdessus, Plenary Meeting of the FATF, pp. 1-4, mimeo.

JOHNSON J., Y.C. D. LIM (2002), Money Laundering: Has the Financial Action Task Force Made a Difference?, Journal of Financial Crime, vol.10, n.1, pp.7-22.

LA PORTA R., LOPEZ-DE-SILANES F., SHLEIFER A., VISHNY R.W. (1998), Law and Finance, Journal of Political Economy, n.106, 1113

MACNEIL, I., (1978) Contracts: Adjustments of Long -Term Economic Relations Under Classical, Neoclassical, and Relational Contract Law, in 72 Northwestern University Law Review, 854.

MADISON, J., (1787) The Federalist n. 10, in The Federalist Papers, J. Cooke eds, New York, 1961.

MAHONEY, P. G., (2000) The Common Law and Economic Growth: Hayek Might be Right, University of Virginia School of Law – Legal Studies Working Papers Series, n. 8.

MASCIANDARO D. (2000) The Illegal Sector, Money Laundering and Legal Economy: A Macroeconomic Analysis, Journal of Financial Crime, n.2, 103-112.

_____________ (1999), Money Laundering: the Economics of Regulation, European Journal of Law and Economics, n.3, May, pp 245 - 240. _____________ (1998), Money Laundering Regulation: the Micro Economics, Journal of Money Laundering Control, Vol.2, n.1, pp. 49 - 58. _____________ (1996), Pecunia Olet? Microeconomics of Banking and Financial Laundering, International Review of Economics and Business, October, pp817-844. MASCIANDARO, D. AND PORTOLANO A. (2002) It Takes Two to Tango: International Financial

Regulation and Off-shore Centres. A Law and Economics Approach, Journal of Money Laundering Control, (forth.)

NORTH, D., (1990) Institutions, Institutional Change, and Economic Performance, Cambridge.

OECD, (1998) Harmful Tax Competition – An Emerging Global Issue, Paris.

OLSON, M., (1965) The Logic of Collective Action, Cambridge.

PERSSON T. and TABELLINI G., (2000) Political Economics: Explaining Economic Policy, Cambridge M.A., MIT University Press.

POSNER, R. A., (1998) Economic Analysis of Law, New York.

RIDER B.A.K. (2002), Weapons of War: the Use of Anti - Money Laundering Laws against Terrorist and Criminal Enterprises, International Journal of Banking Regulation, vol.4, n.1, pp.13-31.

ROGOFF K., (1997) Foreign and Underground Demand for Euro Notes: Blessing or a Curse?, Working Paper, 1997.

Version of 16/01/03 ________________________________________________________________________________________________

35

ROMANO, R., (1999) Corporate Law and Corporate Governance, 365, in Firms, Markets, and Hierarchies, G .R. CARROLL and D. J. TEECE eds, New York.

__________, (1993) The Genius of American Corporate Law, Washington, D.C.

__________, (1985) Law as a Product: Some Pieces of the Incorporation Puzzle, in 1 Journal of Law, Economics, and Organisation, 225.

SINN H.W. and WESTERMANN F., (2001) Why has the Euro been falling?, CESIfo Working Paper No. 493, May.

WASSERMAN, M., (2002) Dirty money, Regional Review, Federal Reserve Bank of Boston, v.12, n.1, pp14-21.

WILLIAMSON, O., (1996) The Mechanisms of Governance, New York.

______________, (1985) The Economic Institution of Capitalism, New York.

______________, (1983) Credible Commitments: Using Hostages to Support Exchange, in 73 The American Economic Review, 519.

______________, (1979) Transaction Cost Economics: The Governance of Contractual Relations, in 22 The Journal of Law and Economics, 233.

YANIV, G. ,(1994) Taxation and Dirty money laundering, Public Finances/Public Finance, 49

(Supplement), pp. 40-51 __________ (1999) Tax Evasion, Risky Laundering, and Optimal Deterrence Policy, International

Tax and Public Finance, 6, pp. 27-38

Version of 16/01/03 ________________________________________________________________________________________________

36

10. FIGURES AND TABLES

FIGURE 1

. E g y p t . Phil ippines

. Indonesia

C o o k Is lands .Marshal l Is lands.

. IsraelL e b a n o n. .Myanmar

.Nauru.

Niue

.Russ ia

.

.Domin ica . St .Vincent & Grenadines.Niger ia

Liechtenstein . .Ukraine

.Grenada. St.Kitts e Nevis.Bahamas

GuatemalaP a n a m a.

.Cayman Islands

.H u n g a r y

Version of 16/01/03 ________________________________________________________________________________________________

37

TABLE 1

Version of 16/01/03 ________________________________________________________________________________________________

38

TABLE 2

0

5

10

15

20

25

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25

Criteria

F r e q u e n c y

Version of 16/01/03 ________________________________________________________________________________________________

39

TABLE 3 : OFF SHORE COUNTRIES Dummy=1 for OFCs, 0 otherwise

Version of 16/01/03 ________________________________________________________________________________________________

40

APPENDIX 1

LIST OF CRITERIA FOR DEFINING NON-COOPERATIVE COUNTRIES OR TERRITORIES A. Loopholes in financial regulations (i) No or inadequate regulations and supervision of financial institutions 1. Absence or ineffective regulations and supervision for all financial institutions in a given country or territory, onshore or offshore, on an equivalent basis with respect to international standards applicable to money laundering. (ii) Inadequate rules for the licensing and creation of financial institutions, including assessing the backgrounds of

their managers and beneficial owners 2. Possibility for individuals or legal entities to operate a financial institution without authorisation or registration or with very rudimentary requirements for authorisation or registration. 3. Absence of measures to guard against holding of management functions and control or acquisition of a significant investment in financial institutions by criminals or their confederates. (ii) Inadequate customer identification requirements for financial institutions 4. Existence of anonymous accounts or accounts in obviously fictitious names. 5. Lack of effective laws, regulations, agreements between supervisory authorities and financial institutions or self-regulatory agreements among financial institutions on identification by the financial institution of the client and beneficial owner of an account: −no obligation to verify the identity of the client; −no requirement to identify the beneficial owners where there are doubts as to whether the client is acting on his own behalf; −no obligation to renew identification of the client or the beneficial owner when doubts appear as to their identity in the course of business relationships; −no requirement for financial institutions to develop ongoing anti-money laundering training programmes. 6. Lack of a legal or regulatory obligation for financial institutions or agreements between supervisory authorities and financial institutions or self-agreements among financial institutions to record and keep, for a reasonable and sufficient time (five years), documents connected with the identity of their clients, as well as records on national and international transactions. 7. Legal or practical obstacles to access by administrative and judicial authorities to information with respect to the identity of the holders or beneficial owners and information connected with the transactions recorded. (iii) Excessive secrecy provisions regarding financial institutions 8. Secrecy provisions which can be invoked against, but not lifted by competent administrative authorities in the context of enquiries concerning money laundering. 9. Secrecy provisions which can be invoked against, but not lifted by judicial authorities in criminal investigations related to money laundering. (iv) Lack of efficient suspicious transactions reporting system 10. Absence of an efficient mandatory system for reporting suspicious or unusual transactions to a competent authority, provided that such a system aims to detect and prosecute money laundering. 11. Lack of monitoring and criminal or administrative sanctions in respect to the obligation to report suspicious or unusual transactions. A. Obstacles raised by other regulatory requirements

Version of 16/01/03 ________________________________________________________________________________________________

41

(i) Inadequate commercial law requirements for registration of business and legal entities 12. Inadequate means for identifying, recording and making available relevant information related to legal and business entities (name, legal form, address, identity of directors, provisions regulating the power to bind the entity). (ii) Lack of identification of the beneficial owner(s) of legal and business entities 13. Obstacles to identification by financial institutions of the beneficial owner(s) and directors/officers of a company or beneficiaries of legal or business entities. 14. Regulatory or other systems which allow financial institutions to carry out financial business where the beneficial owner(s) of transactions is unknown, or is represented by an intermediary who refuses to divulge that information, without informing the competent authorities. B. Obstacles to international co-operation (i) Obstacles to international co-operation by administrative authorities 15. Laws or regulations prohibiting international exchange of information between administrative anti-money laundering authorities or not granting clear gateways or subjecting exchange of information to unduly restrictive conditions. 16. Prohibiting relevant administrative authorities to conduct investigations or enquiries on behalf of, or for account of their foreign counterparts. 17. Obvious unwillingness to respond constructively to requests (e.g. failure to take the appropriate measures in due course, long delays in responding). 18. Restrictive practices in international co-operation against money laundering between supervisory authorities or between FIUs for the analysis and investigation of suspicious transactions, especially on the grounds that such transactions may relate to tax matters. (ii) Obstacles to international co-operation by judicial authorities 19. Failure to criminalise laundering of the proceeds from serious crimes. 20. Laws or regulations prohibiting international exchange of information between judicial authorities (notably specific reservations to the anti-money laundering provisions of international agreements) or placing highly restrictive conditions on the exchange of information. 21. Obvious unwillingness to respond constructively to mutual legal assistance requests (e.g. failure to take the appropriate measures in due course, long delays in responding). 22. Refusal to provide judicial co-operation in cases involving offences recognised as such by the requested jurisdiction especially on the grounds that tax matters are involved. A. Inadequate resources for preventing and detecting money laundering activities (i) Lack of resources in public and private sectors 23. Failure to provide the administrative and judicial authorities with the necessary financial, human or technical resources to exercise their functions or to conduct their investigations. 24. Inadequate or corrupt professional staff in either governmental, judicial or supervisory authorities or among those responsible for anti-money laundering compliance in the financial services industry. (ii) Absence of a financial intelligence unit or of an equivalent mechanism 25. Lack of a centralised unit (i.e., a financial intelligence unit) or of an equivalent mechanism for the collection, analysis and dissemination of suspicious transactions information to competent authorities.

Version of 16/01/03 ________________________________________________________________________________________________

42

APPENDIX 2

CRITERIA VIOLATIONS IN 2000

Data: FATF, www.oecd.org/fatf

Version of 16/01/03 ________________________________________________________________________________________________

43

CRITERIA VIOLATIONS IN 2001

Data: FATF, www.oecd.org/fatf

Version of 16/01/03 ________________________________________________________________________________________________

44

CRITERIA VIOLATIONS IN 2002

Data: FATF, www.oecd.org/fatf